Highway video hierarchical encryption method and system based on traffic abnormal event

Abstract

The application provides a highway video hierarchical encryption method and system based on traffic abnormal events. The method comprises the following steps: collecting highway video data, identifying vehicle feature information and traffic abnormal events, scoring the traffic abnormal events, obtaining an event urgency score and a data correlation score, further processing to obtain a value score, and then obtaining a matched data encryption strategy to encrypt the highway video data, generating a plaintext storage tag containing traffic abnormal event related information, and establishing an index mapping relationship with the encrypted video data. The application realizes differentiated encryption storage strategy by intelligently analyzing the value and sensitivity of the video content, which significantly reduces the system storage and computing overhead while ensuring the security of core data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of video data security, in particular to a highway video hierarchical encryption method and system based on traffic abnormal events. BACKGROUND

[0002] Highway video data is of great value to traffic management, accident identification, public security prevention and control, etc., and its security is crucial. At present, the common practice is to store all video data using a unified strong encryption standard (such as AES-256). However, this approach has certain defects: first, most of the video content is normal traffic pictures without events, and the value density is low, so strong encryption of these videos causes waste of computing and storage resources; second, the unified encryption strategy cannot be differentiated according to the importance of events and the sensitivity of data, and there is a risk of over-protection or insufficient protection; finally, the encrypted data is difficult to quickly search and use, and needs to be decrypted in its entirety when data analysis and evidence retrieval are performed, which is inefficient. In the prior art, although some schemes propose to classify data, they are mostly based on simple rules (such as classification according to camera position), lack dynamic and intelligent evaluation of the value of video content itself, and cannot adapt to the complex and variable actual scene of highways. Therefore, there is an urgent need for an intelligent storage method that can automatically identify the value of video content and implement fine-grained hierarchical encryption accordingly, to optimize resource allocation and improve data utilization efficiency under the premise of ensuring security. SUMMARY

[0003] The purpose of the present application is to provide a highway video hierarchical encryption method and system based on traffic abnormal events, which intelligently and dynamically evaluates the value of video data, and executes differentiated encryption strategies based on the evaluation results, to achieve the coordinated optimization of security protection and resource utilization efficiency, while supporting dynamic upgrading of encryption strategies according to real-time events, providing emergency decryption channels, and introducing a hash digest mechanism to guarantee data integrity, so as to flexibly cope with various complex scenarios and provide reliable data security technical support for intelligent highway construction.

[0004] The present application also provides a highway video hierarchical encryption method based on traffic abnormal events, comprising the following steps:

[0005] Collecting highway video data and identifying vehicle feature information and traffic abnormal events;

[0006] Scoring the traffic abnormal events to obtain an event urgency score and a data relevance score;

[0007] Processing the event urgency score and the data relevance score to obtain a value score, and then obtaining a matched data encryption strategy to encrypt the highway video data;

[0008] Generate a plaintext storage tag containing traffic anomaly event related information, and establish an index mapping relationship with the encrypted video data.

[0009] Optionally, in the highway video hierarchical encryption method based on traffic anomaly events described in the present application, the collection of highway video data and the identification of vehicle feature information and traffic anomaly events comprises:

[0010] Collect highway video data and its basic attribute information, including collection device ID, shooting location data, collection time and video duration;

[0011] Detect vehicles in the video using a target detection algorithm based on deep learning, and identify vehicle feature information;

[0012] Use a preset multi-target tracking algorithm to track the detected vehicles and obtain motion trajectories;

[0013] Based on the motion trajectory, identify traffic anomaly events through a preset spatio-temporal behavior recognition model, and output event type data.

[0014] Optionally, in the highway video hierarchical encryption method based on traffic anomaly events described in the present application, the scoring of traffic anomaly events to obtain event urgency score and data correlation score comprises:

[0015] Score the identified traffic anomaly events using a preset event urgency scoring rule to obtain an event urgency score;

[0016] Extract features associated with traffic anomaly events from highway video, including target association features and spatio-temporal association features;

[0017] Based on the target association features and spatio-temporal association features, use a preset data correlation scoring rule to score the data correlation.

[0018] Optionally, in the highway video hierarchical encryption method based on traffic anomaly events described in the present application, the value score is obtained by processing the event urgency score and data correlation score, and then the matched data encryption strategy is obtained to encrypt the highway video data, comprising:

[0019] According to the event urgency score and data correlation score, weighted calculation is performed to obtain a value score;

[0020] Determine the value level according to the threshold range to which the value score belongs;

[0021] Input the value level into the preset encryption strategy rule library for matching to obtain a data encryption strategy;

[0022] The expressway video data is encrypted according to the data encryption strategy and stored in a retrieval database.

[0023] Optionally, in the expressway video hierarchical encryption method based on traffic abnormal events, the plaintext storage label containing the traffic abnormal event related information is generated, and an index mapping relationship is established with the encrypted video data, including:

[0024] The event type data, value level, vehicle feature information and basic attribute information are generated into a plaintext storage label; a hash digest of the encrypted video data is calculated, the hash digest and the plaintext storage label are stored in a retrieval database, and an index mapping relationship is established with the encrypted video data.

[0025] Optionally, in the expressway video hierarchical encryption method based on traffic abnormal events, the method further includes:

[0026] The traffic abnormal event identification result is monitored in real time, and the event type data is input into a preset emergency event list for comparison;

[0027] If the comparison is successful, an encryption strategy upgrade operation is triggered, the value level is marked as the highest level, and the corresponding encryption strategy is matched;

[0028] Log information of the strategy upgrade is recorded and stored, including an upgrade time point and an event identifier triggering the upgrade.

[0029] Optionally, in the expressway video hierarchical encryption method based on traffic abnormal events, the method further includes:

[0030] The value level change of the same traffic abnormal event is monitored in real time;

[0031] If the promotion number of the value level in a preset time period is greater than or equal to a preset number threshold, the encryption strategy is adjusted to the encryption strategy corresponding to the current highest value level;

[0032] Log information of the strategy upgrade is recorded and stored, including an upgrade time point and an event identifier triggering the upgrade.

[0033] Optionally, in the expressway video hierarchical encryption method based on traffic abnormal events, the method further includes:

[0034] If the value level is the highest level, an emergency decryption instruction is automatically generated;

[0035] The emergency decryption instruction includes basic attribute information, an authorized decryption time window, an authorized rescue unit code, a video segment start time stamp and a duration;

[0036] The emergency decryption instruction is transmitted to a command center platform.

[0037] In a second aspect, the application provides a highway video hierarchical encryption system based on traffic anomaly events, comprising a memory and a processor, wherein the memory stores a program of a highway video hierarchical encryption method based on traffic anomaly events, and the program of the highway video hierarchical encryption method based on traffic anomaly events is executed by the processor to implement the following steps:

[0038] Collecting highway video data and performing vehicle feature information recognition and traffic anomaly event recognition;

[0039] Scoring the traffic anomaly event to obtain an event emergency score and a data correlation score;

[0040] Processing the value score according to the event emergency score and the data correlation score, and then obtaining a matched data encryption strategy to encrypt the highway video data;

[0041] Generating a plaintext storage tag containing traffic anomaly event related information, and establishing an index mapping relationship with the encrypted video data.

[0042] Optionally, in the highway video hierarchical encryption system based on traffic anomaly events provided by the application, the collecting of highway video data and the performing of vehicle feature information recognition and traffic anomaly event recognition comprise:

[0043] Collecting highway video data and its basic attribute information, and the basic attribute information includes a collection device ID, shooting location data, collection time and video duration;

[0044] Detecting vehicles in the video using a target detection algorithm based on deep learning, and recognizing vehicle feature information;

[0045] Tracking the detected vehicles using a preset multi-target tracking algorithm to obtain a motion trajectory;

[0046] Based on the motion trajectory, performing traffic anomaly event recognition through a preset spatio-temporal behavior recognition model, and outputting event type data.

[0047] As can be seen from the above, the highway video hierarchical encryption method and system based on traffic anomaly events provided by the application can intelligently and dynamically evaluate the value of video data, and execute differentiated encryption strategies based on the evaluation results, which can realize the collaborative optimization of security protection and resource utilization efficiency, support dynamic upgrading of encryption strategies according to real-time events, provide emergency decryption channels, and introduce a hash digest mechanism to guarantee data integrity, so as to flexibly cope with various complex scenarios and provide reliable data security technical support for intelligent highway construction.

[0048] Other features and advantages of the present application will be set forth in the following description, and in part will be apparent from the description, or can be learned by practice of the application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly introduced as follows. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0050] Figure 1 The flowchart of the expressway video hierarchical encryption method based on traffic abnormal events provided by the embodiments of the present application;

[0051] Figure 2 The flowchart of the vehicle feature information recognition and traffic abnormal event recognition of the expressway video hierarchical encryption method based on traffic abnormal events provided by the embodiments of the present application;

[0052] Figure 3 The flowchart of the event emergency score and data correlation score acquisition of the expressway video hierarchical encryption method based on traffic abnormal events provided by the embodiments of the present application;

[0053] Figure 4 The flowchart of the matched data encryption strategy acquisition of the expressway video hierarchical encryption method based on traffic abnormal events provided by the embodiments of the present application. DETAILED DESCRIPTION

[0054] The technical solutions of the embodiments of the present application will be described clearly and completely in the embodiments of the present application combined with the drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of the present application.

[0055] It should be noted that similar reference numbers and letters refer to similar items in the following drawings, and therefore, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0056] Please refer to Figure 1 , Figure 1 is a flowchart of a highway video hierarchical encryption method based on traffic abnormal events in some embodiments of the present application. The highway video hierarchical encryption method based on traffic abnormal events is used in terminal equipment, such as computers, mobile terminals, etc. The highway video hierarchical encryption method based on traffic abnormal events includes the following steps:

[0057] S11, collecting highway video data and identifying vehicle feature information and traffic abnormal events;

[0058] S12, scoring traffic abnormal events to obtain event emergency score and data correlation score;

[0059] S13, processing value score according to the event emergency score and the data correlation score, and then obtaining a matched data encryption strategy to encrypt the highway video data;

[0060] S14, generating a clear storage tag containing traffic abnormal event related information, and establishing an index mapping relationship with the encrypted video data.

[0061] It should be noted that the present application realizes the collaborative optimization of security protection and resource efficiency through the complete process of intelligent analysis and identification of highway video data, value quantitative evaluation, differential encryption execution and metadata management and application, specifically including: first, using deep learning target detection, multi-target tracking and spatio-temporal behavior recognition model, the vehicle in the video stream is precisely perceived and behavior analyzed, and the structured traffic abnormal event type data is automatically output, providing high reliability input for subsequent value evaluation. In the value quantitative evaluation stage, a multi-dimensional scoring model is established to integrate event urgency and data correlation. The event threat level and data evidence value are quantitatively calculated by combining pre-set rules and feature extraction, and the data value level is determined according to the weighted scoring result, forming the basis for hierarchical decision-making. Then in the differential encryption execution stage, according to the pre-defined encryption strategy rule library, the value level is mapped to the specific encryption algorithm and strength configuration, and the differential processing from strong encryption to lightweight encryption is performed on the video data, ensuring that high-value data obtains the strongest protection while significantly reducing the processing overhead of low-value data. Finally, in the metadata management and application stage, the key information such as event type, value level and vehicle feature is aggregated and stored in plaintext form, and a bidirectional index mapping relationship is established with the encrypted video data, and a hash digest mechanism is introduced to guarantee data integrity, so as to realize efficient retrieval, multi-dimensional statistical analysis and fast traceability of encrypted video under the premise of ensuring security. The present application also integrates dynamic strategy upgrade and intelligent emergency response mechanism, which can automatically trigger safety strategy adjustment according to real-time traffic event situation, and can effectively improve the active defense and real-time decision-making capability of the system.

[0062] Please refer to Figure 2 , Figure 2 is a flowchart of vehicle feature information identification and traffic abnormal event identification in the highway video hierarchical encryption method based on traffic abnormal events in some embodiments of the present application. According to the embodiment of the present application, the collection of highway video data and the identification of vehicle feature information and traffic abnormal events comprises:

[0063] S21, collecting highway video data and its basic attribute information, the basic attribute information including collection device ID, shooting position data, collection time and video duration;

[0064] S22, detecting vehicles in the video using a deep learning-based target detection algorithm, and identifying vehicle feature information;

[0065] S23, tracking the detected vehicles using a pre-set multi-target tracking algorithm to obtain motion trajectories;

[0066] S24, based on the motion trajectories, identifying traffic abnormal events through a pre-set spatio-temporal behavior recognition model, and outputting event type data.

[0067] It should be noted that, in this embodiment of the application, a preset YOLOv8 model is used to detect vehicles in the video to obtain vehicle feature information (including license plate number, etc.), a preset DeepSORT algorithm is used to track the detected vehicles to obtain motion trajectories, and based on the motion trajectories, a preset spatiotemporal graph convolutional network (ST-GCN) behavior recognition model is used to identify traffic abnormal events and obtain event type data.

[0068] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating the process of obtaining an event urgency score and a data correlation score in a highway video hierarchical encryption method based on traffic anomalies, as described in some embodiments of this application. According to embodiments of the present invention, scoring traffic anomalies to obtain an event urgency score and a data correlation score includes:

[0069] S31. Use the preset event urgency scoring rules to score the identified traffic anomalies and obtain the event urgency score.

[0070] S32. Extract features associated with traffic anomalies from highway videos, including target-related features and spatiotemporal-related features;

[0071] S33. Based on the target association features and spatiotemporal association features, perform data association scoring using preset data association scoring rules.

[0072] It should be noted that, in this embodiment, a preset event urgency scoring rule is used to score the identified traffic anomalies. This scoring rule is based on event type, severity, and potential impact, specifically including: a base score for event type (e.g., traffic accidents, hazardous chemical leaks, and vehicle fires are scored 10 points; severe congestion is scored 6 points; and abnormal parking is scored 3 points); a severity coefficient (e.g., for traffic accidents, the coefficient is set based on the number of vehicles involved: 1.0 for single-vehicle accidents, 1.5 for two-vehicle collisions, and 2.0 for multi-vehicle chain collisions); for congestion, the coefficient is set based on the congestion length (1.0 for congestion length less than or equal to 1 kilometer, 1.2 for congestion length greater than 1 kilometer but less than 5 kilometers, and 1.5 for congestion length greater than or equal to 5 kilometers); and a potential impact coefficient (e.g., 1.5 for congestion occurring at critical nodes such as tunnels and bridges, and 1.0 for congestion occurring on ordinary road sections). Finally, the event urgency score is calculated by multiplying the base score for event type, the severity coefficient, and the potential impact coefficient.

[0073] In this embodiment, the target association feature score is used to evaluate the association and clarity of target objects and events in a video. The feature dimensions include: the size proportion of the target (vehicle, pedestrian) in the image, clarity (whether it is blurry), pose integrity (whether it is occluded), and association strength (whether it is a directly involved vehicle). Examples of scoring rules are as follows: directly involved vehicles (such as vehicles involved in a collision) are assigned 5 points, indirectly related vehicles (such as vehicles swerving to avoid an emergency) are assigned 2 points; a target size proportion greater than 50% is assigned 3 points, 20% < target size proportion ≤ 50% is assigned 2 points, a target size proportion less than or equal to 20% is assigned 1 point, a clearly identifiable license plate / face is assigned 2 points, and severe occlusion or blurriness is assigned 0 points. The target association feature score is the sum of the association and clarity scores.

[0074] In this embodiment, the spatiotemporal correlation feature score evaluates the completeness and viewpoint value of the video capture event process. The feature dimensions include: temporal coverage (whether the entire event process is recorded) and spatial viewpoint (whether the viewpoint is frontal and comprehensive). Examples of scoring rules are as follows: recording a complete process (more than 30 seconds before, during, and more than 60 seconds after the event) is awarded 5 points; recording only the event's occurrence is awarded 3 points; recording only before or after the event is awarded 2 points. A frontal viewpoint (clearly showing the license plate and the vehicle's frontal behavior) is awarded 3 points; a side or top-down panoramic viewpoint (showing the positional relationship between vehicles) is awarded 2 points; and a viewpoint at the edge of the field of view or with a poor viewpoint is awarded 1 point. The spatiotemporal correlation feature score is the sum of the temporal coverage score and the viewpoint value score. The data correlation score is obtained by summing the target correlation feature score and the spatiotemporal correlation feature score.

[0075] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating the process of obtaining a matching data encryption strategy in a highway video tiered encryption method based on traffic anomaly events, as described in some embodiments of this application. According to an embodiment of the present invention, the step of obtaining a value score based on the event urgency score and data correlation score, and then obtaining a matching data encryption strategy to encrypt the highway video data, includes:

[0076] S41. A value score is obtained by weighting the event urgency score and the data relevance score.

[0077] S42. Determine the value level based on the threshold range to which the value score belongs;

[0078] S43. Input the value level into the preset encryption policy rule base for matching to obtain the data encryption policy;

[0079] S44. After encrypting the highway video data according to the data encryption strategy, it is stored in the retrieval database.

[0080] It should be noted that this application employs a refined, multi-dimensional quantitative evaluation-based intelligent video data grading method to meet the real-time requirements of massive video processing on highways. Specifically, a pre-defined encryption strategy rule base predefines the encryption algorithms, encryption strength, and encryption range corresponding to different value levels. For example: Level V4 (extremely high value) uses the national cryptographic standard SM4 or AES-256 algorithm to encrypt the entire video stream (I-frames, P-frames, B-frames); Level V3 (high value) uses the AES-128 algorithm to encrypt keyframes (I-frames); Level V2 (medium value) uses a lightweight encryption algorithm (such as Chacha20) to encrypt the file header and non-critical video data; and Level V1 (low value) can be stored in plaintext or only undergo integrity verification.

[0081] According to an embodiment of the present invention, generating plaintext storage tags containing information related to traffic anomalies and establishing an index mapping relationship with encrypted video data includes:

[0082] The event type data, value level, vehicle characteristic information, and basic attribute information are used to generate plaintext storage tags; the hash digest of the encrypted video data is calculated, and the hash digest and plaintext storage tags are stored in the retrieval database, and an index mapping relationship is established with the encrypted video data.

[0083] It should be noted that by using the analysis results of traffic anomalies as plaintext metadata, calculating the hash digest of the encrypted video data, and storing this hash digest as an integrity verification field along with the plaintext storage tag, while storing it separately from the encrypted video data and establishing an index association, efficient retrieval and secure management of encrypted video data can be achieved. It is worth mentioning that this application calculates and stores the hash value (digital fingerprint) of the encrypted video, and uses hash verification to prove whether the retrieved encrypted video has been modified since storage, providing integrity (tamper-proof) protection for subsequent verification.

[0084] According to an embodiment of the present invention, it further includes:

[0085] The results of the traffic anomaly identification are monitored in real time, and the event type data is entered into a preset emergency event list for comparison.

[0086] If the comparison is successful, the encryption policy upgrade operation is triggered, marking its value level as the highest level and matching the corresponding encryption policy.

[0087] Record and store log information for policy upgrades, including the upgrade time and the event identifier that triggered the upgrade.

[0088] It should be noted that this application provides a dynamic emergency response mechanism that is based on the immediate judgment of event type (compared with the list of emergency events). Once a match is found, the encryption policy is immediately upgraded without waiting for the complete value scoring process. This mechanism is suitable for providing second-level security response to the highest priority events, demonstrating the real-time nature and intelligence of the system.

[0089] According to an embodiment of the present invention, it further includes:

[0090] Real-time monitoring of changes in the value level of the same traffic anomaly;

[0091] If the number of times its value level increases within a preset time period is greater than or equal to the preset level threshold, then its encryption strategy will be adjusted to the encryption strategy corresponding to the current highest value level.

[0092] Record and store log information for policy upgrades, including the upgrade time and the event identifier that triggered the upgrade.

[0093] It should be noted that this application provides a dynamic adjustment mechanism that monitors the level changes of the same event over time. If the level increases significantly in a short period of time (such as when the incident worsens), the encryption strategy is automatically upgraded.

[0094] According to an embodiment of the present invention, it further includes:

[0095] If the value level is the highest, an emergency decryption command will be automatically generated;

[0096] Emergency decryption instructions include basic attribute information, authorized decryption time window, authorized rescue unit code, video clip start timestamp and duration;

[0097] The emergency decryption command is transmitted to the command center platform.

[0098] It should be noted that when the highest level of incident occurs, the system automatically generates emergency instructions with time and scope limitations, and distributes them through the command center platform, enabling controlled and efficient authorized use of data in emergency situations.

[0099] According to an embodiment of the present invention, it further includes:

[0100] Based on the event type, shooting location data, and collection time, the number of abnormal traffic events and the distribution of event types on a specified road segment within a specified time period are statistically analyzed.

[0101] The period when the number of traffic anomalies exceeds a preset threshold is designated as the high-incidence period for accidents, and a statistical report on the high-incidence period for accidents is generated.

[0102] Based on the number of traffic anomalies, event types, and vehicle characteristic information, a curve showing the change in the number of violations for a specific vehicle type within a preset time period is plotted, and a violation trend report for that specific vehicle type is generated.

[0103] It should be noted that this application, through efficient analysis of plaintext metadata, intelligently identifies the high-incidence periods of accidents and the violation trends of specific vehicle types (such as cars, buses, trucks, and hazardous chemical vehicles), and generates intuitive statistical reports, thereby providing key data support for traffic management departments to carry out precise prevention and control and efficient decision-making. The entire process does not require touching the original encrypted data, thus balancing security and practicality.

[0104] This invention also discloses a highway video hierarchical encryption system based on traffic anomaly events, comprising a memory and a processor. The memory stores a program for a highway video hierarchical encryption method based on traffic anomaly events. When the processor executes the program for the highway video hierarchical encryption method based on traffic anomaly events, it performs the following steps:

[0105] Collect highway video data and perform vehicle feature information recognition and traffic anomaly event recognition;

[0106] Traffic anomalies are scored to obtain an event urgency score and a data correlation score;

[0107] A value score is obtained by processing the event urgency score and data correlation score, and then a matching data encryption strategy is obtained to encrypt the highway video data.

[0108] Generate plaintext storage tags containing information related to traffic anomalies and establish an index mapping relationship with encrypted video data.

[0109] It should be noted that this application achieves synergistic optimization of security protection and resource efficiency through a complete process of intelligent analysis and identification, value quantification assessment, differentiated encryption execution, and metadata management and application of highway video data. Specifically, this includes: First, utilizing deep learning object detection, multi-object tracking, and spatiotemporal behavior recognition models, accurate perception and behavior analysis of vehicles in the video stream are performed, automatically outputting structured traffic anomaly event type data to provide highly reliable input for subsequent value assessment. In the value quantification assessment stage, a multi-dimensional scoring model integrating event urgency and data relevance is established. The threat level of the event and the value of data evidence are quantified through a combination of preset rules and feature extraction. Data value levels are then determined based on the weighted scoring results, forming the basis for hierarchical decision-making. Subsequently, in the differentiated encryption execution stage, according to a predefined encryption strategy rule base, the value level is mapped to specific encryption algorithms and strength configurations. Differential processing, ranging from strong encryption to lightweight encryption, is performed on the video data, ensuring that high-value data receives the strongest protection while significantly reducing the processing overhead of low-value data. Finally, in the metadata management and application phase, key information such as event type, value level, and vehicle characteristics are aggregated and stored in plaintext, and a bidirectional index mapping relationship is established with the encrypted video data. A hash digest mechanism is also introduced to ensure data integrity, thereby achieving efficient retrieval, multi-dimensional statistical analysis, and rapid source tracing and evidence collection for encrypted videos while ensuring security. This application also integrates a dynamic policy upgrade and intelligent emergency response mechanism, which can automatically trigger security policy adjustments based on the real-time traffic event situation, effectively improving the system's proactive defense and real-time decision-making capabilities.

[0110] According to an embodiment of the present invention, the step of collecting highway video data and performing vehicle feature information recognition and traffic anomaly event recognition includes:

[0111] Collect video data of highways and its basic attribute information, including the acquisition device ID, shooting location data, acquisition time and video duration.

[0112] Vehicles in videos are detected using deep learning-based object detection algorithms, and vehicle feature information is identified.

[0113] The detected vehicles are tracked using a pre-defined multi-target tracking algorithm to obtain their movement trajectories;

[0114] Based on the motion trajectory, traffic anomaly events are identified using a preset spatiotemporal behavior recognition model, and event type data is output.

[0115] It should be noted that, in this embodiment of the application, a preset YOLOv8 model is used to detect vehicles in the video to obtain vehicle feature information (including license plate number, etc.), a preset DeepSORT algorithm is used to track the detected vehicles to obtain motion trajectories, and based on the motion trajectories, a preset spatiotemporal graph convolutional network (ST-GCN) behavior recognition model is used to identify traffic abnormal events and obtain event type data.

[0116] According to an embodiment of the present invention, the scoring of traffic anomalies to obtain an event urgency score and a data correlation score includes:

[0117] The identified traffic anomalies are scored using preset incident urgency scoring rules to obtain incident urgency scores;

[0118] Extract features associated with traffic anomalies from highway videos, including target-related features and spatiotemporal-related features;

[0119] Based on target association features and spatiotemporal association features, data association is scored using preset data association scoring rules.

[0120] It should be noted that, in this embodiment, a preset event urgency scoring rule is used to score the identified traffic anomalies. This scoring rule is based on event type, severity, and potential impact, specifically including: a base score for event type (e.g., traffic accidents, hazardous chemical leaks, and vehicle fires are scored 10 points; severe congestion is scored 6 points; and abnormal parking is scored 3 points); a severity coefficient (e.g., for traffic accidents, the coefficient is set based on the number of vehicles involved: 1.0 for single-vehicle accidents, 1.5 for two-vehicle collisions, and 2.0 for multi-vehicle chain collisions); for congestion, the coefficient is set based on the congestion length (1.0 for congestion length less than or equal to 1 kilometer, 1.2 for congestion length greater than 1 kilometer but less than 5 kilometers, and 1.5 for congestion length greater than or equal to 5 kilometers); and a potential impact coefficient (e.g., 1.5 for congestion occurring at critical nodes such as tunnels and bridges, and 1.0 for congestion occurring on ordinary road sections). Finally, the event urgency score is calculated by multiplying the base score for event type, the severity coefficient, and the potential impact coefficient.

[0121] In this embodiment, the target association feature score is used to evaluate the association and clarity of target objects and events in a video. The feature dimensions include: the size proportion of the target (vehicle, pedestrian) in the image, clarity (whether it is blurry), pose integrity (whether it is occluded), and association strength (whether it is a directly involved vehicle). Examples of scoring rules are as follows: directly involved vehicles (such as vehicles involved in a collision) are assigned 5 points, indirectly related vehicles (such as vehicles swerving to avoid an emergency) are assigned 2 points; a target size proportion greater than 50% is assigned 3 points, 20% < target size proportion ≤ 50% is assigned 2 points, a target size proportion less than or equal to 20% is assigned 1 point, a clearly identifiable license plate / face is assigned 2 points, and severe occlusion or blurriness is assigned 0 points. The target association feature score is the sum of the association and clarity scores.

[0122] In this embodiment, the spatiotemporal correlation feature score evaluates the completeness and viewpoint value of the video capture event process. The feature dimensions include: temporal coverage (whether the entire event process is recorded) and spatial viewpoint (whether the viewpoint is frontal and comprehensive). Examples of scoring rules are as follows: recording a complete process (more than 30 seconds before, during, and more than 60 seconds after the event) is awarded 5 points; recording only the event's occurrence is awarded 3 points; recording only before or after the event is awarded 2 points. A frontal viewpoint (clearly showing the license plate and the vehicle's frontal behavior) is awarded 3 points; a side or top-down panoramic viewpoint (showing the positional relationship between vehicles) is awarded 2 points; and a viewpoint at the edge of the field of view or with a poor viewpoint is awarded 1 point. The spatiotemporal correlation feature score is the sum of the temporal coverage score and the viewpoint value score. The data correlation score is obtained by summing the target correlation feature score and the spatiotemporal correlation feature score.

[0123] According to an embodiment of the present invention, the step of obtaining a value score based on the event urgency score and data correlation score, and then obtaining a matching data encryption strategy to encrypt highway video data, includes:

[0124] A value score is obtained by weighting the event urgency score and the data relevance score.

[0125] The value level is determined based on the threshold range to which the value score belongs;

[0126] The value level is entered into a preset encryption policy rule base for matching to obtain the data encryption policy;

[0127] Highway video data is encrypted according to a data encryption strategy and then stored in the retrieval database.

[0128] It should be noted that this application employs a refined, multi-dimensional quantitative evaluation-based intelligent video data grading method to meet the real-time requirements of massive video processing on highways. Specifically, a pre-defined encryption strategy rule base predefines the encryption algorithms, encryption strength, and encryption range corresponding to different value levels. For example: Level V4 (extremely high value) uses the national cryptographic standard SM4 or AES-256 algorithm to encrypt the entire video stream (I-frames, P-frames, B-frames); Level V3 (high value) uses the AES-128 algorithm to encrypt keyframes (I-frames); Level V2 (medium value) uses a lightweight encryption algorithm (such as Chacha20) to encrypt the file header and non-critical video data; and Level V1 (low value) can be stored in plaintext or only undergo integrity verification.

[0129] According to an embodiment of the present invention, generating plaintext storage tags containing information related to traffic anomalies and establishing an index mapping relationship with encrypted video data includes:

[0130] The event type data, value level, vehicle characteristic information, and basic attribute information are used to generate plaintext storage tags; the hash digest of the encrypted video data is calculated, and the hash digest and plaintext storage tags are stored in the retrieval database, and an index mapping relationship is established with the encrypted video data.

[0131] It should be noted that by using the analysis results of traffic anomalies as plaintext metadata, calculating the hash digest of the encrypted video data, and storing this hash digest as an integrity verification field along with the plaintext storage tag, while storing it separately from the encrypted video data and establishing an index association, efficient retrieval and secure management of encrypted video data can be achieved. It is worth mentioning that this application calculates and stores the hash value (digital fingerprint) of the encrypted video, and uses hash verification to prove whether the retrieved encrypted video has been modified since storage, providing integrity (tamper-proof) protection for subsequent verification.

[0132] According to an embodiment of the present invention, it further includes:

[0133] The results of traffic anomaly event identification are monitored in real time, and the event type data is entered into a preset emergency event list for comparison.

[0134] If the comparison is successful, the encryption policy upgrade operation is triggered, marking its value level as the highest level and matching the corresponding encryption policy.

[0135] Record and store log information for policy upgrades, including the upgrade time and the event identifier that triggered the upgrade.

[0136] It should be noted that this application provides a dynamic emergency response mechanism that is based on the immediate judgment of event type (compared with the list of emergency events). Once a match is found, the encryption policy is immediately upgraded without waiting for the complete value scoring process. This mechanism is suitable for providing second-level security response to the highest priority events, demonstrating the real-time nature and intelligence of the system.

[0137] According to an embodiment of the present invention, it further includes:

[0138] Real-time monitoring of changes in the value level of the same traffic anomaly;

[0139] If the number of times its value level increases within a preset time period is greater than or equal to the preset level threshold, then its encryption strategy will be adjusted to the encryption strategy corresponding to the current highest value level.

[0140] Record and store log information for policy upgrades, including the upgrade time and the event identifier that triggered the upgrade.

[0141] It should be noted that this application provides a dynamic adjustment mechanism that monitors the level changes of the same event over time. If the level increases significantly in a short period of time (such as when the incident worsens), the encryption strategy is automatically upgraded.

[0142] According to an embodiment of the present invention, it further includes:

[0143] If the value level is the highest, an emergency decryption command will be automatically generated;

[0144] Emergency decryption instructions include basic attribute information, authorized decryption time window, authorized rescue unit code, video clip start timestamp and duration;

[0145] The emergency decryption command is transmitted to the command center platform.

[0146] It should be noted that when the highest level of incident occurs, the system automatically generates emergency instructions with time and scope limitations, and distributes them through the command center platform, enabling controlled and efficient authorized use of data in emergency situations.

[0147] According to an embodiment of the present invention, it further includes:

[0148] Based on the event type, shooting location data, and collection time, the number of abnormal traffic events and the distribution of event types on a specified road segment within a specified time period are statistically analyzed.

[0149] The period when the number of traffic anomalies exceeds a preset threshold is designated as the high-incidence period for accidents, and a statistical report on the high-incidence period for accidents is generated.

[0150] Based on the number of traffic anomalies, event types, and vehicle characteristic information, a curve showing the change in the number of violations for a specific vehicle type within a preset time period is plotted, and a violation trend report for that specific vehicle type is generated.

[0151] It should be noted that this application, through efficient analysis of plaintext metadata, intelligently identifies the high-incidence periods of accidents and the violation trends of specific vehicle types (such as cars, buses, trucks, and hazardous chemical vehicles), and generates intuitive statistical reports, thereby providing key data support for traffic management departments to carry out precise prevention and control and efficient decision-making. The entire process does not require touching the original encrypted data, thus balancing security and practicality.

[0152] The present invention discloses a hierarchical encryption method and system for highway videos based on traffic anomalies. By intelligently and dynamically evaluating the value of video data and implementing differentiated encryption strategies based on the evaluation results, it can achieve synergistic optimization of security protection and resource utilization efficiency. At the same time, it supports dynamic upgrading of encryption strategies according to real-time events, provides an emergency decryption channel, and introduces a hash digest mechanism to ensure data integrity. Thus, it can flexibly cope with various complex scenarios and provide reliable data security technology support for the construction of smart highways.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0154] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0155] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

Claims

1. A hierarchical encryption method for highway video based on traffic anomaly events, characterized in that, Includes the following steps: Collect highway video data and perform vehicle feature information recognition and traffic anomaly event recognition; The identified traffic anomalies are scored using preset incident urgency scoring rules to obtain incident urgency scores; Extract features associated with traffic anomalies from highway videos, including target-related features and spatiotemporal-related features; Target association features represent the degree of association and clarity between target objects and events in a video; Spatiotemporal correlation features represent the completeness of the video capture event process and its perspective value; Based on target association features and spatiotemporal association features, data association is scored using preset data association scoring rules; A value score is obtained by weighting the event urgency score and the data relevance score. The value level is determined based on the threshold range to which the value score belongs; The value level is entered into a preset encryption policy rule base for matching to obtain the data encryption policy; The highway video data is encrypted according to the data encryption strategy and then stored in the retrieval database; Generate plaintext storage tags containing information related to traffic anomalies and establish an index mapping relationship with encrypted video data.

2. The method for hierarchical encryption of highway video based on traffic anomaly events according to claim 1, characterized in that, The process of collecting highway video data and identifying vehicle feature information and traffic anomalies includes: Collect video data of highways and its basic attribute information, including the acquisition device ID, shooting location data, acquisition time and video duration. Vehicles in videos are detected using deep learning-based object detection algorithms, and vehicle feature information is identified. The detected vehicles are tracked using a pre-defined multi-target tracking algorithm to obtain their movement trajectories; Based on the motion trajectory, traffic anomaly events are identified using a preset spatiotemporal behavior recognition model, and event type data is output.

3. The method for hierarchical encryption of highway video based on traffic anomaly events according to claim 2, characterized in that, The process of generating plaintext storage tags containing information related to traffic anomalies and establishing an index mapping relationship with encrypted video data includes: Generate plaintext storage tags from event type data, value level, vehicle characteristic information, and basic attribute information; Calculate the hash digest of the encrypted video data, store the hash digest and plaintext storage tag in the retrieval database, and establish an index mapping relationship with the encrypted video data.

4. The highway video hierarchical encryption method based on traffic anomaly events according to claim 3, characterized in that, Also includes: The results of the traffic anomaly identification are monitored in real time, and the event type data is entered into a preset emergency event list for comparison. If the comparison is successful, the encryption policy upgrade operation is triggered, marking its value level as the highest level and matching the corresponding encryption policy. Record and store log information for policy upgrades, including the upgrade time and the event identifier that triggered the upgrade.

5. The method for hierarchical encryption of highway video based on traffic anomaly events according to claim 4, characterized in that, Also includes: Real-time monitoring of changes in the value level of the same traffic anomaly; If the number of times its value level increases within a preset time period is greater than or equal to the preset level threshold, then its encryption strategy will be adjusted to the encryption strategy corresponding to the current highest value level. Record and store log information for policy upgrades, including the upgrade time and the event identifier that triggered the upgrade.

6. The method for hierarchical encryption of highway video based on traffic anomaly events according to claim 5, characterized in that, Also includes: If the value level is the highest, an emergency decryption command will be automatically generated; Emergency decryption instructions include basic attribute information, authorized decryption time window, authorized rescue unit code, video clip start timestamp and duration; The emergency decryption command is transmitted to the command center platform.

7. A highway video hierarchical encryption system based on traffic anomaly events, characterized in that, The system includes a memory and a processor. The memory stores a program for a highway video hierarchical encryption method based on traffic anomaly events. When the program for the highway video hierarchical encryption method based on traffic anomaly events is executed by the processor, it performs the following steps: Collect highway video data and perform vehicle feature information recognition and traffic anomaly event recognition; The identified traffic anomalies are scored using preset incident urgency scoring rules to obtain incident urgency scores; Extract features associated with traffic anomalies from highway videos, including target-related features and spatiotemporal-related features; Target association features represent the degree of association and clarity between target objects and events in a video; Spatiotemporal correlation features represent the completeness of the video capture event process and its perspective value; Based on target association features and spatiotemporal association features, data association is scored using preset data association scoring rules; A value score is obtained by weighting the event urgency score and the data relevance score. The value level is determined based on the threshold range to which the value score belongs; The value level is entered into a preset encryption policy rule base for matching to obtain the data encryption policy; The highway video data is encrypted according to the data encryption strategy and then stored in the retrieval database; Generate plaintext storage tags containing information related to traffic anomalies and establish an index mapping relationship with encrypted video data.

8. The highway video hierarchical encryption system based on traffic anomaly events according to claim 7, characterized in that, The process of collecting highway video data and identifying vehicle feature information and traffic anomalies includes: Collect video data of highways and its basic attribute information, including the acquisition device ID, shooting location data, acquisition time and video duration. Vehicles in videos are detected using deep learning-based object detection algorithms, and vehicle feature information is identified. The detected vehicles are tracked using a pre-defined multi-target tracking algorithm to obtain their movement trajectories; Based on the motion trajectory, traffic anomaly events are identified using a preset spatiotemporal behavior recognition model, and event type data is output.

Images